CHAPTER

A Simple Procedure to
Analyze Positions of Interest
in Infectious Cell Cultures
by Correlative Light
5
and Electron Microscopy
Kazimierz Madela*, Sebastian Banhart{, Anja Zimmermann{, Janett Piesker*,
Norbert Bannert{, Michael Laue*
*Advanced Light and Electron Microscopy (ZBS 4), Robert Koch Institute, Berlin, Germany
{
Junior Research Group Sexually Transmitted Bacterial Pathogens (NG 5), Robert Koch Institute,
Berlin, Germany
{
Centre for HIV and other Retroviruses (FG 18), Robert Koch Institute, Berlin, Germany

CHAPTER OUTLINE
Introduction and Rationale......................................................................................... 94
5.1 Methods ............................................................................................................96
5.1.1 Cell Culture, Infection and Transfection ............................................ 96
5.1.2 Correlative live-cell Imaging and SEM ............................................... 97
5.1.3 Correlative live-cell Imaging and Thin Section Transmission Electron
Microscopy ..................................................................................... 98
5.2 Materials.........................................................................................................103
5.3 Discussion.......................................................................................................105
Summary ................................................................................................................ 108
Acknowledgments ................................................................................................... 108
References ............................................................................................................. 108

Abstract
Plastic cell culture dishes that contain a thin bottom of highest optical quality including an
imprinted finder grid (m-Dish Grid-500) are optimally suited for routine correlative light and
electron microscopy using chemical fixation. Such dishes allow high-resolution fluorescence
and bright-field imaging using fixed and living cells and are compatible with standard protocols
for scanning and transmission electron microscopy. Ease of use during cell culture and imaging,

Methods in Cell Biology, Volume 124, ISSN 0091-679X, http://dx.doi.org/10.1016/B978-0-12-801075-4.00005-7

© 2014 Elsevier Inc. All rights reserved.
93
94 CHAPTER 5 A Simple Procedure for CLEM

as well as a tight cover render the dishes particularly suitable for working with infectious
organisms up to the highest biosafety level. Detailed protocols are provided and demonstrated
by showing two examples: monitoring the production of virus-like particles of the Human
Endogenous Retrovirus HERV-K(HML-2) by HeLa cells and investigation of Rab11-positive
membrane-compartments of HeLa cells after infection with Chlamydia trachomatis.

INTRODUCTION AND RATIONALE

Cell cultures are one of the most frequently used tools to study infectious organisms,
mainly because cell culture cells are the smallest viable units that can be infected.
Moreover cell cultures are needed to propagate at least some infectious organisms,
like for instance viruses or certain obligate intracellular bacteria, such as Chlamydia.
An important advantage of working with cell cultures is their compatibility with
most of the biosafety conditions which are necessary to avoid unintended infection
of animals and humans. With the automation of light microscopes, live-cell imaging
became a valuable approach for the investigation of infectious organisms. Applica-
tions range from simple observations of the cytopathic effect which a virus is causing
over time to the analysis of virus entry, assembly, and egress using fluorescent virus
proteins (e.g., Brandenburg & Zhuang, 2007; Schudt, Kolesnikova, Dolnik,
Sodeik, & Becker, 2013). Since most infectious organisms are rather small and in
case of viruses usually smaller than the resolution of conventional diffraction-limited
light microscopy, imaging at highest spatial resolution is frequently necessary. Elec-
tron microscopy is able to visualize all infectious organisms known to date without
using specific labeling and is therefore still used as a generic diagnostic tool (Curry,
Appleton, & Dowsett, 2006; Goldsmith et al., 2013). Moreover, electron microscopy
can reveal other, yet unexpected structures associated with infectious organisms
even within the cellular context (which is also coined as the “reference space”;
Brown, Mantell, Carter, Tilly, & Verkade, 2009) and thereby provides a more
complete picture of the ultrastructure than fluorescence microscopy, including all
super-resolution techniques that are currently available. However, to get maximum
information, advantages of the different microscopy modes (i.e., resolution, specific-
ity, dynamics) must be combined. Correlative light and electron microscopy
(CLEM) allows to analyze corresponding positions of interest by light and electron
microscopy in a successive process and to correlate directly the information obtained
by the different microscopy modes.
Many procedures have been described to study cells and other biological systems
by CLEM (compare this volume; Müller-Reichert & Verkade, 2012). In contrast to
other more specialized approaches, we urged to develop a rather universal method,
which can be routinely used to locate positions of interest in infectious cell cultures
by light microscopy for a relocalization in the electron microscope. Importantly,
such a method should be compatible with common biosafety standards up to the
highest biosafety levels (BSL-3 and -4). While rapid freezing methods are rather
complex to handle and most importantly inactivation of microorganisms by the
 Introduction and Rationale 95

complete process (freezing and the follow-up procedures) has still to be assessed in
more detail (Vanhecke, Zuber, Brugger, & Studer, 2012), chemical fixation has
proven that it is stabilizing (Laue, 2010) and inactivating (Rubo, Gardner, &
Webb, 1967) infectious organisms efficiently. Moreover, chemical fixation is easily
applied and usually needs only liquid handling, like it is done for medium exchange.
However, while it is clear that many questions relevant in ultrastructural research can
be answered by using chemical preparation methods (Griffiths, 1993, chapter 2.1),
highest resolution and minimal artifacts can only be achieved by employing cryo-
preparation techniques (e.g., Eltsov & Zuber, 2006).
Since a few years cell culture dishes (m-Dishes; ibidi GmbH) are available for
imaging which are entirely produced out of plastic, including a particular plastic
at the bottom which guarantees highest optical performance in light microscopy.
They are particularly useful for experiments carried out in biosafety laboratories
where glass should be avoided or is not allowed because it can damage personal pro-
tection (e.g., gloves and suit). A variant of such dishes is equipped with a labeled grid
pattern that is imprinted in the bottom allowing localization of positions of interest by
light microscopy. We have selected these dishes (with or without a grid pattern) as
our standard dish for live-cell imaging because they are ready to use (sterile and pre-
treated by charging) and similar in size to other dishes used for conducting cell cul-
ture experiments (e.g., six-well plates) which allows to switch from screening
experiments to correlative microscopy without changing cell culture parameters. An-
other advantage is that the m-Dish plastic is resistant against most chemicals used for
standard preparation of electron microscopy.
In this chapter, we describe a method for correlative imaging of positions of in-
terests in cells cultivated on m-Dishes by using live-cell microscopy followed by ei-
ther ultrathin-section transmission electron microscopy or scanning electron
microscopy (SEM). A few studies already have used the m-Dishes for correlative im-
aging of cells by light and electron microscopy (e.g., Katsen-Globa et al., 2014;
Koppensteiner, Banning, Schneider, Hohenberg, & Schindler, 2012), but without ex-
tensive description of the methodology. We show results from two representative
models as examples to demonstrate feasibility of our protocols. In one model, trans-
fection of cell cultures was used for the generation of virus-like particles (VLPs) of a
reconstituted Human Endogenous Retrovirus type K (HERV-K[HML-2]) (Chudak
et al., 2013; George et al., 2011). A major goal of such experiments was to study
the assembly and release of these viral particles from human cells transfected with
expression plasmids encoding large parts of the reconstituted ancient retroviral ge-
nome. HERV-K(HML-2) proteins and viral particles are expressed in teratocarci-
nomas and several other human cancers and have been associated with
carcinogenesis (Bannert & Kurth, 2006). The VLPs were fluorescent labeled by in-
corporation of a Gag-Cherry fusion protein into the regular Gag lattice during virion
assembly. In pilot experiments, a 1:5 ratio of Gag-Cherry to Gag expression plasmids
was found to give best fluorescent labeling, thus allowing efficient release of mor-
phologically intact VLPs. To achieve a higher expression of the retroviral proteins,
large parts of the coding sequences were codon optimized (George et al., 2011). In
96 CHAPTER 5 A Simple Procedure for CLEM

the other model, cells were infected with Chlamydia trachomatis, a pathogenic sex-
ually transmitted bacterium, that invades human cells and replicates in a membrane-
bound intracellular compartment termed “inclusion” (Moulder, 1991). It is important
to study the formation and maintenance of these inclusions to understand the prop-
agation and pathogenicity of Chlamydia. In particular, reorganization of the mem-
brane compartments, such as the Golgi apparatus (Heuer et al., 2009), and
acquisition of host cell lipids (Hackstadt, Rockey, Heinzen, & Scidmore, 1996;
Rejman Lipinski et al., 2009) are issues for a detailed analysis of cellular trafficking
processes which also involves live-cell imaging using various labeled cell compart-
ments (e.g., Rab11-positive vesicles).

5.1 METHODS
5.1.1 CELL CULTURE, INFECTION AND TRANSFECTION
Culture dishes from ibidi company (m-Dish; 35 mm in diameter, 12 mm high, with
lid) were used for cell culture. Dishes possess a thin (170 mm) plastic bottom in which
a labeled grid was imprinted (Fig. 5.1). Cultivation of cells was performed identically
to standard procedures (six-well plates are perfect for testing of the cultivation con-
ditions, because the individual wells have almost the same size than a m-Dish).
A pretreatment of the plastic surface is usually not necessary, if physically modified
dishes (ibiTreat) are used to promote cell adhesion.
Transfection of Hela cells with HERV-K(HML-2) cDNA constructs
1. HeLa cells were seeded on m-Dishes (150,000 cells/dish) and cultivated in
Dulbecco’s modified Eagle Medium (D-MEM) supplemented with 10% fetal
calf serum (plus 50 units/ml penicillin, 50 mg/ml streptomycin and 2 mM

FIGURE 5.1
Cell culture dish (m-Dish) with imprinted and labeled grid pattern. (A) Overview of the
entire dish (without lid) which is 35 mm in diameter. The effective diameter for imaging is
21 mm, while the grid pattern covers a central area of 10 x 10 mm. Brackets indicate the
central part of the grid pattern which is shown in (B) in detail by using light microscopy. One
field has a size of 500 x 500 mm, i.e., the entire grid pattern is formed by 20 x 20 fields.
 5.1 Methods 97

L-glutamine). Cultivation was done overnight at 37  C, 5% carbon dioxide and

98% humidity.
2. Transfection was performed using PolyFect reagent according to instructions of
the manufacturer and a mixture (1:5) of the plasmids pcDNAoricoGagCherry
(Gag-Cherry) and pcDNAoricoHERV-K_GagProPol (George et al., 2011). After
incubation overnight, reagents were removed and replaced by optimized D-MEM
(Opti-MEM, without phenol red). For live-cell imaging Trolox, a vitamin
E analogue, was added (100 mM) to reduce photo damage by free radicals.

Infection of Hela cells expressing Rab11A/eGFP with Chlamydia

1. HeLa cells were cultured in RPMI 1640 medium supplemented with 10% fetal
calf serum, 1 mM sodium pyruvate, and 2 mM L-glutamine at 37  C and 5% CO2
in a humidified incubator.
2. C. trachomatis was propagated and purified from HeLa cell monolayers.
Infections were performed in D-MEM (with 4.5 g/l glucose) supplemented with
5% fetal calf serum, 1 mM sodium pyruvate, and 2 mM L-glutamine (infection
medium).
3. For live-cell imaging, approximately 100,000 HeLa cells were seeded in a single
m-Dish and cultivated overnight. The next day, cells were washed twice with
infection medium and incubated with C. trachomatis L2 using a multiplicity of
infection of approximately 0.5 at 35  C and 5% CO2 in a humidified incubator.
After 2 h, cells were again washed with infection medium and then incubated at
35  C. At 8 h post infectionem (p.i.), cells were transfected with pEGFP-Rab11A
using Lipofectamine 2000 transfection reagent (Invitrogen GmbH, Germany)
according to the manufacturer’s protocol. Live-cell imaging was performed 24 h
p.i. in infection medium (without phenol red).

5.1.2 CORRELATIVE LIVE-CELL IMAGING AND SEM

Visualization of HERV-K(HML-2) assembly and confirmation of fluorescent
labeled HERV-K(HML-2) VLPs at the surface of HeLa cells
To optimize experimental conditions for studying the assembly and release of
HERV-K (HML-2) VLPs, we have conducted correlative live-cell imaging and
SEM. While live-cell imaging gave us information about the expression and distri-
bution of the fluorescent labeled retrovirus protein (Gag-Cherry) over time, SEM
provided the end-point control for the final formation of VLPs at the cell surface.
The entire cell population was monitored over time by live-cell imaging and individ-
ual cells were selected for a detailed analysis when punctate fluorescence signifi-
cantly concentrated near their cell surface. Position of cells relative to the grid
pattern was noted, overview images were recorded and detailed inspection at higher
magnification followed. After live-cell imaging, cells were immediately fixed
by glutaraldehyde. The dish bottom was removed and processed for SEM. Recorded
positions were easily relocalized relative to the grid pattern at low magnification in
98 CHAPTER 5 A Simple Procedure for CLEM

the SEM and surface of cells were inspected at higher magnification for the presence
of virus particles (Fig. 5.2).
1. Live-cell imaging: Transfected Hela cells were inspected with an automated
live-cell imaging microscope (Nikon Eclipse TE200-E; Nikon Corp., Japan).
Imaging was done using a Plan-Fluor (20x, NA 0.5, DIC) or a Plan-Apochromat
(60x, NA 1.4, DIC, oil) lens and an EMCCD camera (at a resolution of
512  512 pixels, 16 bit) for recording. All fields of interest were inspected
using differential interference contrast (DIC) and fluorescent mode (excitation
filter: 485 nm; emission filter: 526 nm). During imaging cells were maintained
at a temperature of 37  C and in a moistened atmosphere containing 5% carbon
dioxide.
2. Chemical fixation: At the end of the observation period (usually up to a few
hours), cells were fixed by removing the medium and replacing it with the
fixative (2.5% glutaraldehyde in 0.05 M Hepes, pH 7.2). Incubation was at least
for 30 min at 37  C. Usually, samples were stored at 4  C in fixative for a few
days until follow-up preparation for SEM was performed.
3. Sample preparation for SEM: The thin bottom of the dish was extracted by using
a sharp scalpel, avoiding touching the cells and dehydrated with ethanol (30%,
50%, 70%, 90%, 96%, each for 15 min; absolute ethanol, for 60 min, one
change). Drying was performed according to the critical point drying (CPD)
method using carbon dioxide and a critical-point dryer (CPD 030, Bal-Tec,
Liechtenstein). Dried samples were mounted on a sample support (stub) with
adhesive carbon tabs with the cells facing up. Sample surface was coated with a
layer (approximately 5 nm thick) of gold/palladium in a sputter coater.
4. SEM: Imaging was done using a field-emission scanning electron microscope
(Leo 1530 Gemini; Carl Zeiss Microscopy, Germany) at high vacuum
conditions and the in-lens secondary electron detector at 5 kV acceleration
voltage.

5.1.3 CORRELATIVE LIVE-CELL IMAGING AND THIN SECTION

TRANSMISSION ELECTRON MICROSCOPY
Analysis of Rab11/eGFP-positive membrane compartments in Chlamydia
infected HeLa cells
Infection of cells with the pathogenic bacterium C. trachomatis results in the
formation of so-called intracellular inclusions, in which bacteria survive and grow
(Moulder, 1991). Formation and maintenance of the inclusion is accompanied
with a reorganization of the cellular metabolism including membrane trafficking.
In particular, the Golgi apparatus is fragmented into mini-stacks (Heuer et al.,
2009), a process which is depending on functional Rab6 and Rab11 (Rejman
Lipinski et al., 2009). To study this reorganization of membranes in more detail
HeLa cells were infected with Chlamydia and subsequently transfected with
Rab11/eGFP. Expression of Rab11/eGFP and formation of inclusions were observed
in a confocal live-cell imaging microscope (LSM 780). Cells which displayed
 5.1 Methods 99

FIGURE 5.2
Correlative live-cell fluorescence and scanning electron microscopy of HeLa cells which are
producing fluorescently labeled virus-like particles of the Human Endogenous Retrovirus
HERV-K(HML-2). (A) Overlay of images from fluorescence and bright-field light microscopy of
a field labeled “G” in which several cells show red fluorescence from expressed virus fusion
proteins (Gag-Cherry). The inset shows a magnified view of the field marked with a white
rectangle. (B) Scanning electron microscopy (SEM) of the same observation field as shown in
(A). Inset: An increased magnification of the region marked with the rectangle in (A) and (B).
Labeled cells are shown at higher magnification in (C) and (D) or (E) and (F), respectively.
(C–F) SEM of selected cells as indicated in the inset in (B). (C and D) A cell which shows
red fluorescence in (A) and several dot-like particles at the cell surface (C). At higher
magnification (D) particles exhibit uniform size of about 100 nm which corresponds to the
particle size of retroviruses. The inset shows a single budding retrovirus in an ultrathin section
from a different preparation visualized by transmission electron microscopy. (E and F)
Another cell which shows, in contrast to the cell in (C) and (D), no fluorescence in (A) and no
dot-like particles at the cells surface. At higher magnification (F) the cell surface reveals only
typical surface structures that are mainly microvilli of different lengths.
100 CHAPTER 5 A Simple Procedure for CLEM

inclusions and significant Rab11/eGFP expression were analyzed in more detail by

using higher magnification and recording of z-stacks to reveal distribution of fluo-
rescence in 3D. Phase-contrast imaging was used for visualization of inclusions and
bacteria which appeared as granules. Position of cells relative to the grid pattern was
noted and documented by taken overview images. After chemical fixation, dehydra-
tion and embedding, positions of interest were relocalized and trimmed for serial thin
sectioning.
Cells of interest were identified in sections at the transmission-electron micro-
scope using the overview images taken by light microscopy from each position of
interest. Different sections of a series were inspected and cells were photographed
for registration with the respective images from z-stacks recorded with the confocal
microscope. The result allowed a rough correlation of the fluorescent signal pattern
from a z-slice with the structures visible on images taken from serial sections of the
plastic bloc (Fig. 5.4C–E). Regions which showed significant fluorescence were
analyzed in more detail at higher magnification using the transmission electron
microscope. The analysis revealed the presence of small membrane-bound compart-
ments, i.e., vesicles and tubules of different size (Fig. 5.4F), indicating that Rab11/
eGFP positive membrane-bound compartments are recruited to the inclusions at this
stage of infection.
1. Live-cell imaging: Chlamydia-infected and Rab11/eGFP-transfected Hela cells
were investigated with an automated confocal laser-scanning microscope
(LSM 780; Carl Zeiss Microscopy, Germany). Imaging of fluorescence was done
with an excitation wavelength of 488 nm and emission was detected between
500 and 530 nm. Bright-field images were recorded in scanning mode using one
of the photo multipliers (transmitted-light path). All investigations were done
at 35  C and in an atmosphere that was supplemented with 5% carbon dioxide.
The lid of the culture dish was not firmly closed to allow diffusion of gas.
2. Chemical fixation: At the end of the live-cell imaging, culture medium was
removed and cells were fixed either with 4% paraformaldehyde or with 2.5%
glutaraldehyde in 0.05 M Hepes for at least 3 h at room temperature.
Paraformaldehyde-fixed samples were again analyzed by fluorescence
microscopy, before they were postfixed with 2.5% glutaraldehyde in Hepes
buffer. This procedure allowed a rapid arrest of cellular events by
paraformaldehyde fixation and stabilized them for a further analysis by
fluorescence microscopy (e.g., to find regions of interest and to record overview
images) without significant interference with the imaging. The additional
postfixation with glutaraldehyde was necessary to provide a further stabilization
of the cells which is beneficial for preservation of the ultrastructure. Samples
were stored at 4–8  C until further processing.
3. Postfixation and en-bloc staining: All incubations were done at room
temperature, if not stated otherwise. All fluids were removed from the dish and
immediately replaced by fresh fluids. Aldehyde fixatives were removed by
washing twice in Hepes buffer (0.05 M, pH 7.2) for at least 10 min each, before
 5.1 Methods 101

cells were incubated with 1% osmium tetroxide in distilled water for 1 h. After
several washes with distilled water, samples were incubated with tannic acid
(0.1% in 0.05 M Hepes, pH 7.2; for 30 min), followed by two washes (10 min
each) with sodium sulfate (Na2SO4, 1%) and distilled water (3 for 10 min
each). Block contrasting with uranyl acetate (2% in distilled water) was
performed for 1 h.
4. Dehydration and embedding: Dehydration of samples was performed in ethanol
(70%, 80%, 90%, 100%, 100% ethanol; 15 min each). Infiltration with the resin
(Epon) started with incubation in acetone as an intermedium (3  for 15 min
each) followed by mixtures of acetone and resin (3:1, 1:1, 1:3; for 1 h each) and
pure resin (overnight). Finally, resin was replaced for fresh resin and samples
were cured in an oven at 60  C for at least 24 h. Note that the m-Dish is not
compatible with propylene oxide as intermedium.
5. Relocation of positions: Polymerized samples were inspected with an inverted
light microscope (Nikon Eclipse TE2000-E) using a Plan-Fluor 20x (NA 0.5,
DIC M/N2) lens and standard bright field imaging or an upright microscope
(Zeiss Axiophot) using a Plan-Neofluar 10x (NA 0.3, Ph1) and phase-contrast
imaging. Recorded positions were identified and inspected if cells were still
present or not (Figs. 5.3 and 5.4A and B). Positions of interest were roughly
labeled with a fine marker pen. The mark was used to label cutting lines for
extraction of the respective regions by using a saw or a motorized cutting tool.
Extracted plastic blocs were directly mounted into a holder of the ultramicrotome
or onto a “dummy” plastic bloc.

FIGURE 5.3
Correlative live-cell fluorescence and transmission electron microscopy. Comparison of the cell
locations before and after embedding in resin (Epon) to check the presence of relevant cells and
to select particular positions. (A) Overlay of fluorescence and differential interference contrast
mode images to show positions of HeLa cells, infected with Chlamydia trachomatis, and
expression of Rab11/eGFP (recorded with the Nikon Eclipse TE2000-E). (B) Phase contrast
image of corresponding region after embedding in resin.
102 CHAPTER 5 A Simple Procedure for CLEM

FIGURE 5.4
Correlative live-cell fluorescence and transmission electron microscopy (TEM) of HeLa cells
which expressed Rab11/eGFP and which were infected with Chlamydia trachomatis.
(A) Confocal fluorescence microscopy of the Rab11/eGFP signal (overlay with bright-field
transmission channel). (B) Relocalization of the same cells in the resin-embedded sample
by using light microscopy and bright-field imaging. (C) Confocal fluorescence microscopy
of a relevant region of one cell localized in the center of (A). The fluorescent signal from
Rab11/eGFP shows a distinct distribution between the three inclusions (*) visible in the
cytoplasm. (D) Ultrastructure of the region corresponding to (C) in ultrathin sections
visualized by TEM. (E) Overlay of the manual registration of fluorescence and ultrastructure
that is shown in (C) and (D). (F) Ultrastructural detail of the boxed region in (E) which
shows a variety of small membrane compartments that are visible between two of the
inclusions. Chlamydia (c) are localized close to the bounding membrane of the inclusions.
 5.2 Materials 103

FIGURE 5.5
Ultrathin sections taken parallel to the bottom surface of an embedded m-Dish. (A) A ribbon of
sections (arrows) is floating on the water surface of the knife boat (blue (appearing dark grey
in the lower part of the image in print version)). (B) Magnified view of a section which clearly
shows the grid pattern and a label (“K”) in the upper part that corresponds to the thin plastic of
the dish bottom. The lower part of the section (*) corresponds to the resin above the dish
bottom which contains the cells. Note that the inscriptions are partially visible by transmission
electron microscopy in the very first sections above the dish bottom.

6. Ultramicrotomy: Under the dissection microscope, sample blocs were trimmed

laterally to a width of approximately 1 mm keeping regions of interest in the
middle of the bloc face. The plastic of the dish bottom was removed either by
using a precise motorized sample trimming tool or the ultramicrotome and a
diamond for semithin sectioning. With the latter approach maximum control
of the depth of plastic removal was achieved. The transition between the dish
and the plastic embedded sample was indicated by appearance of the grid
pattern in the sections that were swimming on the trough of the diamond knife
(Fig. 5.5). Serial ultrathin sections (60–80 nm thickness) were collected on
copper slot grids which are covered by a film of pioloform.
7. Transmission electron microscopy: Sections were inspected in a transmission
electron microscope (Tecnai12 BioTwin) at 120 kV without section staining
or occasionally after staining sections with 2% uranyl acetate and 0.1% lead citrate.
8. Image registration: Registration of z-images from confocal fluorescence
microscopy with images from transmission electron microscopy of physical sections
was done manually using the Photoshop software (Adobe Inc., USA). Shape and
distribution pattern of inclusions present in the cytoplasm facilitated registration.

5.2 MATERIALS
Cell culture
1. Cells: HeLa cervical epithelial cells (American Type Culture Collection [ATCC],
No. CCL-2).
2. Bacteria: C. trachomatis Lymphogranuloma venereum biovar strain L2/434/Bu
(ATCC No. VR-902B).
104 CHAPTER 5 A Simple Procedure for CLEM

3. Cell culture dishes (m-Dish): 35 mm in diameter, with grid pattern (500 mm pitch;
m-Dish 35 mm, high Grid-500, sterile, ibiTreat; ibidi GmbH, Germany). Note:
m-Dishes with glass bottom and grid pattern (500 and 50 mm pitch) are also
available. To allow proper imaging by DIC, particular lids which possess a glass
insert (DIC-lid) were used. For further details compare information provided by
ibidi GmbH on their website (www.ibidi.com).
4. Medium, according to the cells used: Dulbecco’s modified Eagle Medium
(D-MEM; Gibco/Invitrogen GmbH, Germany); RPMI 1640 medium
(Gibco/Invitrogen GmbH, Germany); Opti-MEM I (reduced serum, without
phenol red; Gibco/Invitrogen GmbH, Germany).

Infection and transfection

1. HERV-K expression plasmids: Plasmid pcDNAoricoHERV-K_GagProPol has
been described previously (George et al., 2011). It allows CMV-promoter driven
expression of reconstituted original Gag, Gag-Pro and Gag-Pro-Pol proteins
of HERV-K(HML-2). These proteins do assemble into VLPs that bud from the
cell surface (George et al., 2011). Plasmid pcDNAoricoGag-Cherry encodes
the Gag with a C-terminally fused Cherry protein (Chudak et al., 2013).
2. RAB11 expression plasmid: RAB11A transcript variant 1 ligated into pEGFP-C1
(pEGFP-RAB11A; Clontech Laboratories, Inc., USA).
3. Transfection reagents: PolyFect (Qiagen GmbH, Germany); Lipofectamine 2000
(Invitrogen GmbH, Germany).

Live-cell imaging
1. Nikon Eclipse TE2000-E: Motorized microscope (Nikon Corp., Japan) with
inverted optics and a Xenon illumination system (175 W) that is able to rapidly
switch emission wave length (Lambda DG4, Sutter Instrument Company, USA)
for epi-fluorescence imaging. Installed optical contrast modes: phase and DIC.
Image recording was done using a color CCD-camera (DS-2Mv; Nikon Corp.,
Japan) or an EMCCD camera (Cascade 512B; Roper Scientific, USA). The
system was equipped with an automatic focusing system (Perfect-Focus System,
T-PFS; Nikon Corp., Japan) and a cultivation chamber that was heated and
supplied with moistened artificial air (supplemented with 5% carbon dioxide).
2. Zeiss LSM 780: Motorized and digitally controlled confocal laser scanning
microscope with inverted optics (Axio Observer Z1; Carl Zeiss Microscopy
GmbH, Germany) that was equipped with three laser systems: a diode laser
(405 nm emission line), a multiple-line Argon-laser (458, 488, 514 nm), a
HeNe-laser (543, 594, 633 nm). Detection of fluorescence was done using two
parallel photomultipliers or a 32 channel gallium arsenide spectral detector
(QUASAR GaAsP 32). Bright-field images were recorded in scanning mode
using one of the photo multipliers (transmitted-light path). Sample and optics
were surrounded by a cultivation chamber (XL S1 LSM 710; Carl Zeiss
Microscopy GmbH, Germany) that was heated and supplied with carbon dioxide
(final concentration: 5%).
 5.3 Discussion 105

Scanning electron microscopy

1. Fixative: glutaraldehyde (25%; TAAB Laboratories Equipment Ltd., United
Kingdom).
2. Critical point dryer (CPD 030, Bal-Tec, Liechtenstein) supplied with liquid
carbon dioxide.
3. Sputter coater (E5100, Polaron, United Kingdom) equipped with an Au/Pd
target.
4. Field-emission scanning electron microscope, Leo 1530 Gemini (Carl Zeiss
Microscopy, Germany).

Thin section transmission electron microscopy

1. Fixatives: Paraformaldehyde (Merck, Germany); glutaraldehyde (25%; TAAB
Laboratories Equipment Ltd, United Kingdom); osmium tetroxide (Carl Roth,
Germany); tannic acid (e.g., Electron Microscopy Services, No. 21710).
2. Epon resin: 23.52 g glycidyl ether 100, 12.35 g dodenyl succinic anhydride,
14.13 g nadic methyl anhydride, 0.65 g dimethyl phthalate (Serva,
Electrophoresis, Heidelberg, Germany).
3. Ultramicrotome (e.g., UC7; Leica Microsystems, Germany).
4. Sample trimming tool (EM RAPID; Leica Microsystems, Germany).
5. Diamond knife (e.g., Histo or Ultra 45 ; Diatome, Switzerland).
6. Section and en bloc staining: uranyl acetate (2% in distilled water; e.g., Merck,
Germany); lead citrate (e.g., 0.1% in distilled water; Serva, Germany).
7. Transmission electron microscope (e.g., Tecnai12 BioTwin; FEI Corp., The
Netherlands), equipped with a digital CCD camera (MegaViewIII; Olympus Soft
Imaging Solutions, Germany) operated at a resolution of 1376 x 1014 pixels,
14 bit.

5.3 DISCUSSION
Cell culture dishes that possess a plastic bottom of highest optically quality and an
imprinted grid pattern (m-Dishes) for localization of interesting positions are highly
suitable for routinely performed CLEM. They are delivered ready-to-use and usually
can be applied without any change of the cultivation conditions which are used for
cultivation of cells in other plastic containers (e.g., a six-well plate). The m-Dishes
can be easily fitted to most light microscopes and stable live-cell imaging can be
performed. After analysis by light microscopy the m-Dishes allow relocalization
of positions of interest for convenient imaging by either routine scanning or trans-
mission electron microscopy.
CLEM using SEM has the advantage of combining two techniques with a compa-
rable large field of view. This provides representative information of a sample and fa-
cilitates detection of rare events or structures. Several applications may profit from this
application, e.g., the proof of pathogen uptake or release in a subset of cells. CLEM using
SEM provides a very economical approach to check many positions in a single sample,
106 CHAPTER 5 A Simple Procedure for CLEM

which increases throughput and improves data quality by increasing the number of cells/
samples inspected. SEM, in most cases, is restricted to information collected from out-
side of the cells and intracellular events are usually not detectable. However, using
uncoated or slightly carbon-coated samples, internal structures may be visible by tuning
the voltage from low voltages (surface information) to higher voltages (depth informa-
tion) and/or energy filtering of the backscattered electrons if structures of interest con-
tain heavy metals (Rohde, Branitzki-Heinemann, & Talay, 2013).
Different cultivation systems or procedures for CLEM using SEM were pub-
lished before. Most of them use particular culture dishes with a glass bottom
(e.g., from MaTek Corp.) or coverslips that possess particular grids or labels at their
surface (e.g., Larson, Johnson, Webb, & Vogt, 2005). While glass is usually optically
perfect and resistant against many chemicals, removal from the dish may be cumber-
some. Coverslips have to be inserted in a dish or container for live-cell imaging
which might impair with the imaging due to diffraction at the different layers. An-
other interesting procedure for CLEM in conjunction with SEM is to integrate a fluo-
rescence microscope in the SEM (Liv et al., 2013) or to attach it to the SEM chamber
(ClairScope; Morrison et al., 2012). However, both procedures need particular ma-
chines and either do not allow fluorescence imaging of living cells or use a rather
unconventional SEM imaging of the material contrast which provides also informa-
tion from inside the cells (ClairScope).
Intracellular structures are optimally visualized by using thin section electron mi-
croscopy. Still, the most commonly used instrument for analyzing ultrathin section is
the transmission electron microscope. The major problem in CLEM using thin sec-
tions is the proper registration of positions acquired by the different imaging modes.
While in many cases, a formal image registration is not needed, because only the
position of the cell is relevant, in some cases a perfect match of positions is neces-
sary. A couple of factors affect registration of images and especially in 3D, accurate
registration is difficult to achieve (compare discussion and procedures in Kukulski,
et al., 2012 and Padman, Bach, & Ramm, 2014). The image registration accuracy
using the m-Dish and procedures described in this chapter is not precisely known.
It is definitely not better than the resolution of the light microscope which is around
250 nm in xy and much more in z. With confocal imaging of fluorescence a depth
resolution of around 500 nm and above, depending on the pinhole size of the micro-
scope, is expected, which is rather low in comparison to the depth resolution of a
conventional ultrathin section (60–80 nm). Thus a perfect match between images
obtained with both modes, even if serial optical and physical sectioning is applied,
never would be achievable. As a consequence, in the example demonstrated, Rab11/
eGFP-labeled vesicles could not be distinguished reliably from nonlabeled vesicles
which are localized in close vicinity (Fig. 5.4E and F). However, immunogold label-
ing directed against the fluorescent protein provides a possibility to correlate local-
ization of fluorescence signals with ultrastructural details at highest resolution
(e.g., Mironov & Beznoussenko, 2013; Spiegelhalter et al., 2010). The m-Dish seems
to be compatible with pre-embedding and postembedding immunogold labeling. Pre-
liminary experiments showed, that low-temperature embedding in Lowicryl K4M
 5.3 Discussion 107

(HM20 dissolves the bonding of the thin bottom to the plastic dish) allows postem-
bedding immunolabeling for the green fluorescence protein (GFP) (M. Laue, unpub-
lished data).
The CLEM procedure described in the present paper relies on chemical fixation of
the cells, because high-pressure freezing of cells cultivated on the dish plastic did not
result in a sufficient freezing quality (Laue, unpublished observations), most probably
because thermal conductivity of the plastic is not good enough. In addition, working
with infectious pathogens needs to follow particular biosafety rules, which are in many
cases not compatible with high-pressure freezing, because pathogens may be acciden-
tally aerosolized during sample handling or the freezing process. While this must be
avoided with the highest possible likelihood for pathogens of the risk group 3 and 4
(pathogens that are causing diseases with a significant mortality that are extremely dif-
ficult to treat or untreatable), pathogens belonging to the risk group 2 (pathogens that
cause diseases with no significant mortality and available medical treatment) might be
fixed by high-pressure freezing. However, diseases that are caused by pathogens be-
longing to risk group 2 are in some cases far from being harmless and high-pressure
freezing is not a sterilization procedure (compare Vanhecke et al., 2012). Researchers
should consider the individual risks of their procedures and contact their local safety
officers for further support to avoid spread of infections.
Stabilization of cells by chemical fixation is well established and is easily per-
formed by a simple exchange of fluids within the culture dish, which can be done
in a biosafety cabinet to increase biosafety. Inactivation properties of formaldehyde
and glutaraldehyde are well documented for many pathogens (Rubo et al., 1967;
Sabel, Hellman & McDade, 1969). In case of uncertainty, sterility controls can be
performed which rules out that samples are still infectious, which is important to
prove if samples must be transferred to laboratories of a lower biosafety class
(e.g., an electron microscopy laboratory). Besides advantages for biosafety, stabili-
zation of cell cultures by chemical fixation provides flexibility. At the end of an ex-
periment, cell cultures can be routinely fixed and stored for a considerable time until
decision of the follow up-processing, whether it might be electron microscopy or an-
other light microscopy technique. To conserve fluorescence, initial formaldehyde
fixation can be performed, as we did in some of the experiments with Chlamydia.
Those samples can be reinspected under the light microscope after storage (e.g.,
to record more positions of interest) before starting with the follow-up preparation
including further fixation steps to preserve structures for electron microscopy.
The major drawback of chemical fixation is that it is slow and incomplete which
causes changes of the ultrastructure. In case of live-cell imaging experiments a con-
siderable delay in the arrest of cellular events has to be expected in addition. In those
cases, morphology and fluorescent signals should be checked after fixation. Prefixa-
tion with formaldehyde, as indicated earlier, and reinspection of cells of interest
might be a way to solve this issue. If chemical fixation is not sufficient, high-pressure
freezing might be applied, considering all of the biosafety aspects as discussed ear-
lier. In many cases, a suitable alternative would be to work with attenuated or inac-
tive variant of the infectious pathogen (e.g., Romero-Brey et al., 2012).
108 CHAPTER 5 A Simple Procedure for CLEM

Alternative culture systems suitable for CLEM of chemically fixed cell cultures
have been described in many papers before (e.g., Jiménez et al., 2009; Mironov &
Beznoussenko, 2013). Most of them would allow to get the same information as
obtained with the methods described in this chapter. However, the procedures de-
scribed in this chapter most likely are much easier performed and more flexible than
most of the other procedures described before which helps to analyze more samples
in shorter time. Glass or plastic supports, like cover slips or Aclar, that are somehow
labeled for recording positions of interest have to be inserted in culture containers for
imaging which usually impairs high resolution imaging because of inappropriate
thickness or optical properties (compare Padman et al., 2014). Moreover, they
may move during live-cell imaging, if not fixed in the containers. Glass supports usu-
ally are more resistant toward chemicals, but may cause problems during or after
embedding (e.g., separation of glass from resin bloc may impair cells).

SUMMARY
The m-Dishes with an imprinted finder grid provide many advantages for CLEM:
ready-to-use culture dish of a standard size; highest optical performance for fluores-
cence and bright-field imaging of living and fixed cells; compatibility with SEM and
thin section transmission electron microscopy using chemical fixation and standard
follow-up procedures; control of presence of cells at various steps of the preparation;
compatibility with biosafety regulations up to the highest biosafety-level. Hence,
CLEM using m-Dishes is suited to answer routinely a variety of important basic ques-
tions relevant for the study of infectious microorganisms, for example, to search cell
cultures for infected cells by live-cell imaging followed by electron microscopy to
prove the presence of the pathogen within an individual cell, or, to investigate the
ultrastructure of an infected cell at the end of a live-cell imaging experiment that
has followed either fluorescent labeled pathogens or molecules related to infection.

ACKNOWLEDGMENTS
We would like to thank Christoph Schaudinn and Thomas Müller-Reichert for their valuable
comments.
Declaration: The authors declare the absence of competing interests.

REFERENCES
Bannert, N., & Kurth, R. (2006). The evolutionary dynamics of human endogenous retroviral
families. Annual Review of Genomics and Human Genetics, 7, 49–173.
Brandenburg, B., & Zhuang, X. (2007). Virus trafficing—Learning from single-virus tracking.
Nature Reviews. Microbiology, 5, 197–208.
 References 109

Brown, E., Mantell, J., Carter, D., Tilly, G., & Verkade, P. (2009). Studying intracellular trans-
port using high-pressure freezing and correlative light and electron microscopy. Seminar
in Cell & Developmental Biology, 20, 910–919.
Chudak, C., Beimforde, N., George, M., Zimmermann, A., Lausch, V., Hanke, K., et al.
(2013). Identification of late assembly domains of the human endogenous retrovirus-
K(HML-2). Retrovirology, 10, 140.
Curry, A., Appleton, H., & Dowsett, B. (2006). Application of transmission electron micros-
copy to the clinical study of viral and bacterial infections: Present and future. Micron, 37,
91–106.
Eltsov, M., & Zuber, B. (2006). Transmission electron microscopy of the bacterial nucleoid.
Journal of Structural Biology, 156, 246–254.
George, M., Schwecke, T., Beimforde, N., Hohn, O., Chudak, C., Zimmermann, A., et al.
(2011). Identification of the protease cleavage sites in a reconstituted Gag polyprotein
of an HERV-K(HML-2) element. Retrovirology, 8, 30.
Goldsmith, C. A., Ksiazek, T. G., Rollin, P. E., Comer, J. A., Nicholson, W. L., Peret, T. C. T.,
et al. (2013). Cell culture and electron microscopy for identifying viruses in diseases of
unknown cause. Emerging Infectious Diseases, 19, 886–891.
Griffiths, G. (1993). Fine structure immunocytochemistry. Berlin, Heidelberg: Springer.
Hackstadt, T., Rockey, D. D., Heinzen, R. A., & Scidmore, M. A. (1996). Chlamydia tracho-
matis interrupts an exocytic pathway to acquire endogenously synthesized sphingomyelin
in transit from the Golgi apparatus to the plasma membrane. EMBO Journal, 15, 964–977.
Heuer, D., Rejman Lipinski, A., Machuy, N., Karlas, A., Wehrens, A., Siedler, F., et al. (2009).
Chlamydia causes fragmentation of the Golgi compartment to ensure reproduction.
Nature, 457, 731–735.
Jiménez, E. G., Van Donselaar, E. G., De Winter, D. A. M., Vocking, K., Verkleij, A. J., &
Post, J. A. (2009). Gridded Aclar: preparation methods and use for correlative light and
electron microscopy of cell monolayers, by TEM and FIB-SEM. Journal of Microscopy,
237, 208–220.
Katsen-Globa, A., Meiser, I., Petrenko, Y. A., Ivanov, R. V., Lozinsky, V. I., Zimmermann, H.,
et al. (2014). Towards ready-to-use 3-D scaffolds for regenerative medicine: Adhesion-
based cryopreservation of human mesenchymal stem cells attached and spread within
alginate-gelatin cryogel scaffolds. Journal of Material Science: Materials in Medicine,
25, 857–871.
Koppensteiner, H., Banning, C., Schneider, C., Hohenberg, H., & Schindler, M. (2012).
Macrophage internal HIV-1 is protected from neutralizing antibodies. Journal of Virology,
86, 2826–2836.
Kukulski, W., Schorb, M., Welsch, S., Picco, A., Kaksonen, M., & Briggs, J. A. G. (2012).
Precise, correlated fluorescence microscopy and electron tomography of Lowicryl sections
using fluorescent fiducial markers. Methods in Cell Biology, 111, 235–257.
Larson, D. R., Johnson, M. C., Webb, W. W., & Vogt, M. V. (2005). Visualization of retrovirus
budding with correlated light and electron microscopy. Proceedings of the National Acad-
emy of Sciences of the United States of America, 102, 15453–15458.
Laue, M. (2010). Electron microscopy of viruses. Methods in Cell Biology, 96, 1–20.
Liv, N., Zonnevylle, A. C., Narvaez, A. C., Effting, A. P. J., Voorneveld, P. W., Lucas, M. S.,
et al. (2013). Simultaneous correlative scanning electron and high- NA fluorescence
microscopy. PLoS One, 8, e55707.
Mironov, A. A., & Beznoussenko, G. V. (2013). Correlative microscopy. Methods in Cell
Biology, 113, 209–254.
110 CHAPTER 5 A Simple Procedure for CLEM

Morrison, I. E. G., Dennison, C. L., Nishiyama, H., Suga, M., Sato, C., Yarwood, A., et al.
(2012). Atmospheric scanning electron microscope for correlative microscopy. Methods
in Cell Biology, 111, 307–324.
Moulder, J. W. (1991). Interaction of Chlamydiae and host cells in vitro. Microbiological
Reviews, 55, 143–190.
Müller-Reichert, T., & Verkade, P. (2012). Correlative light and electron microscopy.
Amsterdam: Elsevier (Methods in Cell Biology 111).
Padman, B. S., Bach, M., & Ramm, G. (2014). An improved procedure for subcellular spatial
alignment during live-cell CLEM. PLoS One, 9, e95967.
Rejman Lipinski, A., Heymann, J., Meissner, C., Karlas, A., Brinkmann, V., Meyer, T. F., et al.
(2009). Rab6 and Rab11 regulate Chlamydia trachomatis development and golgin-84-
dependent Golgi fragmentation. PLoS Pathogens, 5, e1000615.
Rohde, M., Branitzki-Heinemann, K., & Talay, S. (2013). Kinetic studies in Streptococcal
pathogenesis applying field emission scanning electron microscopy (FESEM). In
R. Rachel (Ed.), Proceedings of the microscopy conference 2013, Regensburg. Part II
(pp. 282–283).
Romero-Brey, I., Merz, A., Chiramel, A., Lee, J.-Y., Chlanda, P., Haselman, U., et al. (2012).
Three-dimensional architecture and biogenesis of membrane structures associated with
Hepatitis C virus replication. PLoS Pathogens, 8, e1003056.
Rubo, S. D., Gardner, J. F., & Webb, R. L. (1967). Biocidal activities of glutaraldehyde and
related compounds. Journal of Applied Bacteriology, 30, 78–87.
Sabel, F., Hellman, A., & McDade, J. J. (1969). Glutaraldehyde inactivation of virus in tissue.
Applied Microbiology, 17, 645–646.
Schudt, G., Kolesnikova, L., Dolnik, O., Sodeik, B., & Becker, S. (2013). Live-cell imaging of
Marburg virus-infected cells uncovers actin-dependent transport of nucleocapsids over
long distances. Proceedings of the National Academy of Sciences of the United States
of America, 110, 14402–14407.
Spiegelhalter, C., Tosch, V., Hentsch, D., Koch, M., Kessler, P., Schwab, Y., et al. (2010).
From dynamic live cell imaging to 3D ultrastructure: Novel integrated methods for high
pressure freezing and correlative light-electron microscopy. PLoS One, 5, e9014.
Vanhecke, D., Zuber, B., Brugger, S. D., & Studer, D. (2012). Safe high-pressure freezing of
infectious micro-organisms. Journal of Microscopy, 246, 124–128.